Neural network (or parallel distributed processing) models have been shown to have some potential for solving optimisation problems. Most formulations result in NP-complete problems and solutions rely on energy based ...
详细信息
Neural network (or parallel distributed processing) models have been shown to have some potential for solving optimisation problems. Most formulations result in NP-complete problems and solutions rely on energy based models, so there is no guarantee that the network converges to a global optimal solution. In this paper, we propose a non-energy based neural shortest path network based on the principle of dynamic programming and least take all network. No problem of local minima exists and it guarantees to reach the optimal solution. The network can work purely in an asynchronous mode which greatly increases the computation speed.
A control system architecture is described for an autonomous flying vehicle. The vehicle, equipped with fourteen sensors, uses a model helicopter as an airframe. The control system utilizes these sensors to: remain al...
详细信息
A control system architecture is described for an autonomous flying vehicle. The vehicle, equipped with fourteen sensors, uses a model helicopter as an airframe. The control system utilizes these sensors to: remain aloft and in stable flight; navigate to a target; and manipulate a physical object. The overall approach to the problem is based on a behavioral paradigm. The key contribution of this paper is the demonstration of a situated agent under these severe circumstances; as the craft is airborne, it is in constant risk of crashing. Unlike terrestrial mobile robots, the craft must constantly make sound decisions to maintain its integrity.
Previously, the authors gave planning algorithms for the problem of grasping and orienting planar parts using a parallel-jaw gripper, with analysis simplified by assuming zero friction between the part and the gripper...
详细信息
Previously, the authors gave planning algorithms for the problem of grasping and orienting planar parts using a parallel-jaw gripper, with analysis simplified by assuming zero friction between the part and the gripper. Their analysis is generalized to allow Coulomb friction. Coulomb friction introduces ranges of orientations where a part can become wedged between two points on its boundary. It is shown that for any planar part operating under Coulomb friction, there is a part (linear/circular in boundary) operating under zero friction that has the same mechanical behavior. Based on this observation, a bound on the coefficient of friction is given that permits the existence of a plan to orient a given part. For friction below this bound, a new planning algorithm that will find a plan for orienting the part is described. It is shown that this plan works under the weaker assumption that friction introduces nondeterminism into the transfer function, i.e., there can be more than one possible outcome for certain ranges of grasp angles.< >
The registration mark problem is defined, as follows. Given an n-sided rigid planar polygonal part and a set of k planar geometric transforms, locate a point on the surface of the part that maximizes the minimum dista...
详细信息
The registration mark problem is defined, as follows. Given an n-sided rigid planar polygonal part and a set of k planar geometric transforms, locate a point on the surface of the part that maximizes the minimum distance between transformed points. A registration mark at this point will be maximally robust to sensor imperfections. An O(n log/sup 2/k+k/sup 4/ log k log*k) time algorithm for solving this planar problem is given, and the result is demonstrated using the Image-Calc vision system. The results extend to curved planar parts and polyhedral parts. How part shape can be redesigned to simplify localization is considered. In essence, it is shown how to design the surface of a part to facilitate automated assembly.< >
Most neural network models are energy based models and there is no guarantee that the network converges to a global optimal solution. A new neural shortest path network model is proposed in which no special convergenc...
详细信息
Most neural network models are energy based models and there is no guarantee that the network converges to a global optimal solution. A new neural shortest path network model is proposed in which no special convergence procedure needs to be performed. The network can work in a purely asynchronous mode, and is guaranteed to reach the global optimal solution.< >
A new concept for navigation of mobile robots based on the planary coded environment (PLACER) is presented, which enables the coordination of several mobile robots in a real manufacturing environment. It is a path ind...
详细信息
A new concept for navigation of mobile robots based on the planary coded environment (PLACER) is presented, which enables the coordination of several mobile robots in a real manufacturing environment. It is a path independent external navigation system with passive environment structuring over the entire movement area of the robots. This area is completely covered with tiles, each containing the coordinates of its centre point inscribed using an error correcting code
Our objective is to recognize parts in a structured environment (such as a factory) using inexpensive and widely available hardware. In this paper we consider the class of parts with constant polygonal cross section. ...
详细信息
The authors present some results on a theoretical semantic device called the checklist paradigm which gives the theoretical bounds on the performance of particular many-valued implication operators and other connectiv...
详细信息
The authors present some results on a theoretical semantic device called the checklist paradigm which gives the theoretical bounds on the performance of particular many-valued implication operators and other connectives. In its most general form, the checklist paradigm pairs the distinct connectives of the same logical type to provide the bounds for interval-valued approximate inference. The global structure imposed on the many-valued connectives by the certain type of checklist paradigm contracting measures is shown to be the S/sub 2*2*2/ group.< >
This paper critically studies approaches that use the theory of fuzzy sets for multi objective decision making with linear programming. Two main approaches are distinguished: one that converts the problem into a singl...
详细信息
An approach for tracing, representation, and recognition of a handwritten numeral in an offline environment is presented. A 2D spatial representation of a numeral is first transformed into a 3D spatiotemporal represen...
详细信息
An approach for tracing, representation, and recognition of a handwritten numeral in an offline environment is presented. A 2D spatial representation of a numeral is first transformed into a 3D spatiotemporal representation by identifying the tracing sequence based on a set of heuristic rules acting as transformation operators. Given the dynamic information of the tracing sequence, a multiresolution critical-point segmentation method is proposed to extract local feature points, at varying degrees of scale and coarseness. A neural network architecture, the hierarchically self-organizing learning (HSOL) network (S. Lee, J.C. Pan, 1989), especially for handwritten numeral recognition, is presented. Experimental results based on a bidirectional HSOL network indicated that the method is robust in terms of variations, deformations, and corruption, achieving about 99% recognition rate for the test patterns.< >
暂无评论